Gene annotations in model organisms such as Drosophila are important contributors to our understanding of the functions of human genes, including human disease-associated genes. Paralogs, which share a common ancestor, present a challenging case for identification of gene function in model organisms as in some cases, loss of function of one paralog is masked by compensatory function of the other(s), such that only when both are disrupted will informative phenotypes be observed. In other cases, redundancy is partial, such that knockout of one paralog has a subset of the phenotypes observed when more than one member of the group is disrupted simultaneously. Recent advances in CRISPR technology by our group and others makes it now possible to systematically knockout paralog pairs in Drosophila, including in a stage- and tissue-specific manner. We will use our existing infrastructure for bioinformatics-based identification of orthologs and paralogs, efficient large-scale production of fly stocks, and fly stock and data sharing to develop a resource useful for double-knockdown of paralogs. Our initial characterization of the genes with regards to signal transduction and neurodegeneration, as well as in-depth analyses by the community, will uncover function for paralogous genes, helping to close the ?phenotype gap? (lack of associated loss-of-function phenotypes) that currently exists for nearly half of all genes in this important genetic model system. The result will be a fly stock resource for further study by Drosophila experts, a bioinformatics pipeline and methods that can be applied to other model systems, and a data resource that will inform annotation of fly and human genes